From: SPERFETTO (sperfetto@hiv.hjf.org)
Date: Thu Dec 28 1995 - 09:04:14 EST
Lorie,
In a recent exp. to determine the concentration and the biological activity of
the CD4 receptors which would bind purified gp120 we performed a competitive
binding exp. with serial dilution's of gp120. After which we added anti-Leu3a
(which competes for the same site ) and OKT3 (which does not compete). These
means were than calculated into ABC's (Quantum beads by FCS) and graphed.
Cumulative frequency calculation are based on a accurate negative control which
may mean blocking with "cold" ( non-conjugated) MAB followed by the same
conjugated MAB. Because isotypic controls may have a different binding affinity
this should not be used into these exps. unless you can prove the same binding
affinity.
I hope the above exp. will give some idea how I would do your exp.
Stephen P. Perfetto
Department of HIV Disease Prevention
Walter Reed Army Institute of Research
1600 East Gude Drive
Rockville, MD. 20850
_______________________________________________________________________________
Subject: Histogram Statistics
From: lori_m@SMTPGATE.BCSEW.EDU at Internet_Gateway
Date: 12/27/95 1:31 PM
Happy New Year!
I'm posting for a colleague with a statistics question:
The basic question is this... If you overlay two histograms and do a
K/S statistical analysis, it will give you a % probability that the
curves are different. This has been used to show that binding in the
experiment has occurred. If you want to show that binding has NOT
occurred or has been blocked, at what level of probability can you say
that the two curves are NOT different?
How do you prove both that binding has occurred (two curves are
different) and that you have now been able to block binding (two
curves are not different)?
Are people out there using other statistics to show this? What type
of numbers show statistical significance in your results?
I haven't seen much in publications about interpretation of data like
this. Feel free to reply directly to me or post back to the group.
Thanks in advance for the help!
Lorie
_____________________________________________________________________
Lorie Miller
FACS Core Lab
Blood Center of Southeastern Wisconsin
lori_m@smtpgate.bcsew.edu
This archive was generated by hypermail 2.1.6 : Thu Jan 01 2004 - 17:30:41 EST
